From mboxrd@z Thu Jan 1 00:00:00 1970 From: Julien Grall Subject: Re: [PATCH 3/3] xen: arm: correctly handle removing a subset of a superpage mapping. Date: Tue, 15 Jul 2014 13:13:54 +0100 Message-ID: <53C51B02.80504@linaro.org> References: <1405355194.31863.2.camel@kazak.uk.xensource.com> <1405355225-4623-3-git-send-email-ian.campbell@citrix.com> <53C422A0.5070405@linaro.org> <1405417133.9188.5.camel@kazak.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1405417133.9188.5.camel@kazak.uk.xensource.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Ian Campbell Cc: stefano.stabellini@eu.citrix.com, tim@xen.org, xen-devel@lists.xen.org List-Id: xen-devel@lists.xenproject.org On 15/07/14 10:38, Ian Campbell wrote: > On Mon, 2014-07-14 at 19:34 +0100, Julien Grall wrote: >> Hi Ian, >> >> On 07/14/2014 05:27 PM, Ian Campbell wrote: >>> This can be exercised for example via ballooning which will remove 4K >>> regions fromanywhere in the address space. >> >> Missing space between "from" and "anywhere" >> >>> >>> Reported-by: Julien Grall >>> Signed-off-by: Ian Campbell >>> --- >>> xen/arch/arm/p2m.c | 29 +++++++++++++++++++++++++++-- >>> 1 file changed, 27 insertions(+), 2 deletions(-) >>> >>> diff --git a/xen/arch/arm/p2m.c b/xen/arch/arm/p2m.c >>> index a10cbaf..68a19bd 100644 >>> --- a/xen/arch/arm/p2m.c >>> +++ b/xen/arch/arm/p2m.c >>> @@ -604,8 +604,33 @@ static int apply_one_level(struct domain *d, >>> return P2M_ONE_PROGRESS_NOP; >>> } >>> >>> - if ( level < 3 && p2m_table(orig_pte) ) >>> - return P2M_ONE_DESCEND; >>> + if ( level < 3 ) >>> + { >>> + if ( p2m_table(orig_pte) ) >>> + return P2M_ONE_DESCEND; >>> + >>> + if ( op == REMOVE && >>> + !is_mapping_aligned(*addr, end_gpaddr, >>> + 0, /* maddr doesn't matter for remove */ >>> + level_size) ) >>> + { >>> + /* >>> + * Removing a mapping from the middle of a superpage. Shatter >>> + * and descend. >>> + */ >>> + *flush = true; >>> + rc = p2m_create_table(d, entry, >>> + level_shift - PAGE_SHIFT, flush_cache); >>> + if ( rc < 0 ) >>> + return rc; >>> + >>> + p2m->stats.shattered[level]++; >>> + p2m->stats.mappings[level]--; >>> + p2m->stats.mappings[level+1] += LPAE_ENTRIES; >>> + >>> + return P2M_ONE_DESCEND; >>> + } >>> + } >> >> I would put the code below (which is only for level3) in the else part. >> This will also remove now useless "if ( level == 3 )". > > This would break the relinquish case, I think. > > Although thinking about it perhaps relinquish should behave like remove > and shatter. My original thinking was that we only ever relinquish the > entire address space so we shouldn't really be hitting that case. Oh right, I though there was a return on every part of this if. Sorry for the noise. With the minor change in the commit message: Acked-by: Julien Grall Regards, -- Julien Grall